FanPost

Points Per Pass Metric, 2010 Season (All QBs)

Welcome to the fourth annual edition of my passing efficiency results. For those unfamiliar with the statistical details, see below. It's the same explanation I've posted since 2007.

After Romo's sterling 2009 campaign--the best of his career--he took a step backwards last year, although he still performed well enough to land in the league's top ten. Of particular interest is his interceptions rate. After improving every single year he's been a starting quarterback, last year he fell back to a rate of 3.3%, close to his 2008 rate. I am not greatly concerned about this apparent setback, because of the normal variation that occurs from year to year. Another way of saying this is he probably was worse than 2009 would lead us to believe, but better than his showing last season. The fact Romo also had to play games when trailing a good portion of the time certainly did not help. The other frequently discussed factor to explain Romo's decline is, of course, the offensive line. It's been beaten to death here, and while I agree it hindered Romo's performance last season, the evidence I've seen is most fans tend to overestimate its importance.

A more relevant discussion, in my opinion, is his age. My research indicates Romo is in a phase where he is and will be a good quarterback for several more years, but his performance will gradually be in decline. As much as anything else, his performance in 2010 is probably evidence of that reality.

So how long will Romo continue to be effective? That's hard to answer with confidence. Originally I used this passing efficiency metric to determine if, based on Romo's first season as a starter, he was a one-year wonder or someone special. After comparing his performance at the age of 26 to the most-similar group of historical NFL quarterbacks at the same age, I concluded he would be a very good quarterback. When I take that same group of similar quarterbacks and follow their career trajectories, their starting careers lasted until they were between 31 and 41 years old. That's not much useful information. If I exclude the two most-similar quarterbacks from the 1960's on the grounds that conditioning and medical care improved significantly since then, we're still left with the same range of 31 to 41, but the remaining group is more clearly distributed: Scott Mitchell 31, Jim Everett 34, Fouts 37, Elway 38, Favre 41 and counting. My sense is Romo is in the middle of that group, so I would say barring a catastrophic injury he should be effective until 36. You heard it here first, sports fans!

Jon Kitna turned in a solid but unspectacular season in relief of Romo. I appreciate his leadership, dedication and courage...but he's no Tony Romo. Thanks for stepping in and helping right a badly listing ship. Let's only hope he doesn't have to get into any games next year.

Interestingly, Stephen McGee performed slightly above the historical norms for 25-year olds, although his limited action, mainly against the second-string Eagle defense, is probably not representative of his true performance. I am encouraged that he did not throw an interception in 44 attempts.

Background on my Point Per Pass Metric:

I've been using a passing offense metric for many years now to gauge how well a QB is doing. Inspired by baseball SABRmetrician Bill James, I performed linear regressions of NFL statistics and concluded from a big picture point of view, only two statistics mattered: Yards gained from scrimmage, and turnovers. The best fit to matching points scored came when assuming the following coefficients:

Ten yards gained from scrimmage = +1.0 points; and
One turnover lost = -6.0 points.

Simply put, teams that score points rack up yardage and don't commit turnovers. Note how powerful turnovers are: A team can, for example, gain 55 yards of offense, but if they turn it over the next play, they actually ended up hurting their team.

With that in mind, I created the points per pass metric, which calculates the effectiveness of quarterbacks. I consider it a better measure of quarterback performance than the official NFL rankings, since it weighs the value of fumbles, and has direct ties to points scored. The above two coefficients are used to determine, on average, how many points a quarterback (and his offensive teammates, since everyone else contributes to his success or failure) creates with each passing attempt. It takes the simple yards per attempt metric we're all familiar with, adjusts it for sacks, and further adjusts it for turnovers. One interception equates to minus six points; one fumble equates to minus three points. I treat fumbles as half a turnover, since there's essentially a 50/50 chance of recovering or losing a fumble. The results will probably be unsurprising for the most part, but sometimes they can appear odd. One of the things that may cause these oddities are fumbles. Most people, when looking over a QB's stat line, don't think about fumbles; they just look at yardage, attempts, touchdowns, interceptions, and maybe sacks. But fumbles are very important, since they are essentially half a turnover.

As with last season, I included not only the raw points per pass value, but also the normalized value, by dividing the raw result by the league average. Therefore, an average quarterback will have a normalized value of 1.00, and an above average quarterback will have a value greater than 1.00. A rigorous statistician would have done research to make sure the mean or average is an appropriate normalizing factor—as opposed to the median, quartile or something else—but time has been at a premium for me this year, so the mean will have to do for now.

One final note. This will be the final season I use the -6.0 and +0.1 coefficients. When I first ran the regressions back in the mid-1980's, I didn't have either the analytical computing power or the academic background in statistics I do now. After I finish grad school this fall, I will rerun my regressions using statistical software and follow rigorous methods, such as transforming into log or other space if needed, ensuring the residuals are normally distributed, and so on. It could yield not only new coefficients but potentially new variables. Stay tuned for next year.

Notes on the rest of the league:

  • The league average last season rose to 0.409, the highest in league history, narrowly surpassing the 0.408 of 2008, and as I predicted rebounding nicely after dipping to 0.397 in 2009. Passing efficiency has been climbing steadily since the late 1970's, and I believe this long-term trend will continue for the foreseeable future. The pass-happy climate has led to efficiency inflation, which was why I went to a normalization process last year. It makes comparisons between years more meaningful.
  • Rookie watch: Four rookies threw 100 or more passes last season: Sam Bradford, Jimmy Clausen, Colt McCoy and John Skelton. Relative to historic performance for their age, Bradford and Skelton performed very well, McCoy was about average and Clausen does not appear to be an NFL quarterback. The most surprising performance was by Tim Tebow, who I flatly doubted would ever be able to play quarterback well in the NFL. His value of 1.15 was the only rookie performance to be above league average, for those rookies that threw at least 50 passes. Count me as still in the skeptic's camp, but he does bear watching in the future.
  • Tom Brady bounced back from his merely excellent 2009 season to rank as the most efficient quarterback last season. However, like Peyton Manning, he is in decline, and with several of the young quarterbacks starting to emerge, I expect him to slide out of the top five next season.
  • Last season I predicted Rivers would rise to be the best quarterback in the league for the next several years, but he'll have to wait until next year to wear the crown, finishing second behind Brady. Regardless of the actual standings, he's the best of the prime-age quarterbacks in the league, something I noticed a couple of years ago. His great performance in the face of a mediocre offensive line, after losing his best receiver (Vincent Jackson) and seeing his great but aging tight end get another year older speaks volumes of how great quarterbacking is much more important than the surrounding cast.
  • I also was championing Aaron Rogers a few years ago, and now he's blossomed into one of the very best in the league.
  • Roethlisberger appears to be resurrecting himself, improving his performance for the second straight year and clearly emerging from his 2006 to 2008 slump. He's almost at the level of his first two seasons, which if he can continue to improve would put him back in the discussion of the league's best.
  • Peyton turned in his worst season since 2002. From 2003 to 2006, he put together the best four season stretch in NFL history. I say this to anyone who will listen: Peyton Manning, even when adjusting for the inflated passing stats of the current era, is the greatest quarterback ever. Now at 35, he is merely an excellent quarterback. Ironically, Eli has never in his career had a season as good as Peyton's 1.22 last season.
  • Speaking of Eli, another tough season for him. He's been below the average value for his age in four of his seven seasons, and I think it's clear he's basically an average quarterback.
  • For those who recall my discussions from previous years, I noticed Drew Brees alternated excellent seasons with not-so-good years, and mistakenly thought he would drop in 2009. Instead, he turned in the second-best season of his career. My guess last season was he was due to regress, and boy did he ever, falling to a very average value. However, I fully expect him to bounce back strong in 2011 and once again be one of the best quarterbacks in the league--I'm not going to bet against the pattern until I see him completely break it.
  • Michael Vick, never a good NFL quarterback, suddenly transformed into one of the league's best last season. His mirror image was Donovan McNabb, who always performed very well for the Eagles--he is easily the best quarterback in franchise history--but who looked very average last season for the Redskins. Personally, I would be a little worried if I were the Eagles. Can Andy Reid bet on Vick in his thirties, inevitably when quarterbacks start to decline, to be the answer to lead the Eagles to the Super Bowl? The answers beyond Vick are not encouraging, particularly since they are so eager to unload Kevin Kolb.
  • Young quarterbacks watch: Matt Ryan continues to perform above average for a player of his age, and is someone we need to keep our eyes on. Many quarterbacks take a significant step forward at the age of 26. On the other hand, it's hard to make sense of Josh Freeman's 2010 campaign. His rookie season was horrible, but last year he ended up in the top five. The truth almost certainly lies somewhere in between. Stay tuned. Similarly, Joe Flacco, who got off to a quick start as a rookie, appears to be treading water. He's still a good quarterback, but his career trajectory seems to be flattening out. On the other hand, Mark Sanchez's two seasons has been utterly average for a quarterback of his age, remarkable since the Jets have one of the very best offensive lines in the league.

Regular Starting Quarterbacks:

Name, Team PPP Normalized
Brady, NE 6.57 1.60
Rivers, SD 6.03 1.47
Roethlisberger, Pit 5.87 1.43
Rogers, GB 5.83 1.42
Freeman, TB 5.29 1.29
Vick, Phi 5.22 1.27
Schaub , Hou 5.20 1.27
Romo, Dal 5.20 1.26
Orton, Den 5.17 1.26
Cassel, KC 5.09 1.24
Manning, Ind 5.03 1.22
Ryan, Atl 4.86 1.18
Flacco, Bal 4.65 1.13
Favre, Min 4.27 1.04
Brees, NO 4.17 1.01
Smith, SF 4.11 1.00
Hill, Det 4.07 0.99
Campbell, Oak 3.95 0.96
Kitna, Dal 3.94 0.96
Manning, NYG 3.92 0.95
Sanchez, NYJ 3.88 0.94
Palmer, Cin 3.85 0.94
Collins, Ten 3.79 0.92
McNabb, Was 3.74 0.91
Fitzpatrick , Buf 3.69 0.90
McCoy, Cle 3.57 0.87
Henne, Mia 3.53 0.86
Bradford, StL 3.46 0.84
Cutler, Chi 3.43 0.83
Hasselbeck, Sea 3.37 0.82
Anderson, Ari 3.15 0.77
Garrard, Jax 3.14 0.76
Kolb, Phi 2.44
0.59
Clausen, Car 1.58
0.39

Quarterbacks with less than 160 and at least 30 attempts:

Name, Team PPP Normalized
Young, Ten 4.82 1.17
Wallace, Cle 4.78 1.16
Tebow, Den 4.75 1.15
Smith, SF 3.95 0.96
McGee, Dal 3.89 0.95
Stafford, Det 3.79 0.92
Stanton, Det 3.74 0.91
Henne, Mia 3.36 0.82
Flynn, GB 3.36 0.82
Grossman, Was 3.15 0.77
Skelton, Ari 2.87 0.70
Gradkowski, Oak 2.83 0.69
Bouman, Jax 2.74 0.67
Dixon, Pit 2.51 0.61
Whitehurst, Sea 2.44 0.59
Webb, Min 2.26 0.55
Delhomme, Cle 2.02 0.49
Batch, Pit 1.72 0.42
Edwards, Jax 1.28 0.31
Jackson, Min 0.44 0.11
Moore, Car 0.30 0.07
Smith, Ten -0.98 -0.24
Hall, Ari -2.50 -0.61

Quarterbacks with less than 30 attempts:

Name, Team PPP Normalized
Pennington, Mia 9.50 2.31
Volek, SD 8.00 1.95
McCown, Jax 6.32 1.54
Johnson, TB 5.72 1.39
Hanie, Chi 4.44 1.08
Hoyer, NE 4.13 1.01
Leftwich, Pit 3.89 0.95
Brunell, NYJ 3.57 0.87
Redman, Atl 3.33 0.81
Palko, KC 3.13 0.76
Clemens, NYJ 3.00 0.73
Pike, Car 2.85 0.69
Bartel, Ari 2.60 0.63
St. Pierre, Car 0.97 0.24
Carr, SF 0.54 0.13
Palmer, Cin -0.40 -0.10
Croyle, KC -2.22 -0.54
Brohm, Buf -4.19 -1.02
Chase, NO -6.00 -1.46
Collins, Chi -8.52 -2.07
Brown, Buf -12.00 -2.92
Boller, Oak -14.20 -3.45

Another user-created commentary provided by a BTB reader.